Loop through array

You can loop over items in an awk array a using the syntax for (i in a). So for example you could do something like

awk '{split($NF,a,","); $NF=""; for (i in a) sub(i"=",a[i]"=",$0); print}' list.txt

If there are always 3 items (x=y) and no space in the lines, this awk statement should work:

awk -F',| |=' '{printf "%s %s %s %s %s=%s %s=%s %s=%s\n", \
$1, $2, $3, $4, $11, $6, $12, $8, $13, $10}' list.txt

Explanation:

• -F',| |=': set the field separator to ,, space and =
• '{printf ... print the values in the desired format
• list.txt the input file

This script should work for any number of items in a row.

while read line do valType=($(echo $line | awk '{ print $NF }' | tr ',' ' ')) vals=($(echo $line | awk '{ $NF="";$1="";$2="";$3="";$4=""; print $0 }' | tr [1234567890] ' ' | tr '=' ' ' )) echo $line | awk '{print $1" "$2" "$3" "$4 }' | tr -d '\n' numVals=${#vals[@]} for i in $(seq 0 $((numVals-1))) do echo -n " "${valType[$i]}"="${vals[$i]} done echo " " done < list.txt

Explanation:

valType contains an array of the comma seperated elements.

vals contains the associated values between the 4th and last positions in the row, with numbers and equals signs removed.

Looping through the number of values in the arrays, the matching values in both arrays can be referenced.

set '.\(=.[^ ]*\) ' '\([^,]*\),'
sed "s/\$/,/;s/$1$1$1$2$2$2/\4\1 \5\2 \6\3/
" <<\DATA
List of animals SET 1=Dog 2=Crow 3=Snake Pet,bird,reptile
List of Countries SET 1=France 2=Singapore 3=Columbia Europe,Asia,SouthAmerica
List of Parts SET 1=KeyBoard 2=Speaker 3=LEDpanel Computer,AudioPlayer,Television
List of Cities SET 1=Amsterdam 2=KualLumpur 3=Paris Netherlands,Malaysia,France
DATA

That uses the shell array to set your array.

OUTPUT

List of animals SET Pet=Dog bird=Crow reptile=Snake
List of Countries SET Europe=France Asia=Singapore SouthAmerica=Columbia
List of Parts SET Computer=KeyBoard AudioPlayer=Speaker Television=LEDpanel
List of Cities SET Netherlands=Amsterdam Malaysia=KualLumpur France=Paris